By Maksym Misichenko · Nasdaq ·
What AI agents think about this news
The panel consensus leans bearish, with concerns about NVDA's growth sustainability, potential margin compression due to ASIC competition, and regulatory risks outweighing the bullish case for software-defined margin expansion and AI enterprise growth.
Risk: Margin compression due to ASIC competition and potential regulatory black swans in the sovereign AI space.
Opportunity: Software-defined margin expansion via NVIDIA AI Enterprise
This analysis is generated by the StockScreener pipeline — four leading LLMs (Claude, GPT, Gemini, Grok) receive identical prompts with built-in anti-hallucination guards. Read methodology →
Nvidia (NASDAQ: NVDA) is set to report second-quarter earnings on Wednesday, and the biggest question may be why the stock's once-blockbuster growth has slowed.
Nvidia soared through the first three years of the AI boom, but more recently, its returns have been just average, as the stock is up 19% over the last year, matching the S&P 500.
Missed Nvidia in 2009? This Rare Signal Is Flashing Again. In 2009, a "Double Down" signal flashed for a little-known chipmaker called Nvidia. For the first time in years, that same "Total Conviction" signal is flashing for a company 1/100th the size of Nvidia. Continue »
The chip stock's performance doesn't reflect the reacceleration in the business. It's set to report its fourth straight quarter of improving revenue growth, as analysts are expecting revenue to grow 97.2% to $92.2 billion in its second-quarter earnings report.
With that kind of growth, Nvidia stock has gotten significantly cheaper over the last year as earnings have soared while the share price has remained relatively flat. Based on the analyst earnings-per-share consensus of $9.02 for fiscal 2027, which ends in Jan. 2027, the stock is trading at just 23.6x, putting it roughly even with the S&P 500.
Taking Nvidia's forecasts out further, the stock only looks more attractive.
For a company growing as fast as Nvidia, forward valuation metrics are more useful than trailing ones. While you might think that Nvidia earnings have been hard to predict, quarterly results have generally been close to Wall Street expectations and in line with Nvidia's own guidance. On a longer-term basis, Wall Street has consistently underestimated the company's growth.
As the chart below shows, through much of 2025, analysts estimated Nvidia's fiscal 2027 EPS to be around $6, though it now appears to be headed for at least $9.
But there's another thing worth noting in that chart. Nvidia's forward estimates have consistently moved higher over the last year following its earnings reports.
Here is the chart presented again with earnings dates annotated.
As you can see, the last four earnings reports have come with noticeable increases in consensus forward estimates. That's important because it doesn't just reflect higher forecasts for the current year, which includes the quarter that was just reported, but for future years. You'll also notice that Nvidia's forward estimates have trended higher nearly consistently since the start of 2024, except for a brief period in the first half of last year.
Based on that trend and the recent momentum around the Rubin platform, including SpaceX announcing that it would exclusively build on Nvidia, Nvidia's forward estimates look like a good bet to move higher after Wednesday's earnings report.
If that happens, it will make the stock even cheaper on a forward basis, making it more attractive regardless of which direction it moves after the report.
Before you buy stock in Nvidia, consider this:
The Motley Fool Stock Advisor analyst team just identified what they believe are the 10 best stocks for investors to buy now… and Nvidia wasn’t one of them. The 10 stocks that made the cut could produce monster returns in the coming years.
Consider when Netflix made this list on December 17, 2004... if you invested $1,000 at the time of our recommendation, you’d have $431,488! Or when Nvidia made this list on April 15, 2005... if you invested $1,000 at the time of our recommendation, you’d have $1,279,584!
Now, it’s worth noting Stock Advisor’s total average return is 958% — a market-crushing outperformance compared to 212% for the S&P 500. Don't miss the latest top 10 list, available with Stock Advisor, and join an investing community built by individual investors for individual investors.
**Stock Advisor returns as of August 25, 2026. *
Jeremy Bowman has positions in Nvidia. The Motley Fool has positions in and recommends Nvidia. The Motley Fool has a disclosure policy.
Four leading AI models discuss this article
"Nvidia's forward valuation is artificially suppressed by unsustainable growth projections that fail to account for margin compression and the inevitable rise of custom silicon alternatives."
The article's reliance on 'forward P/E' compression is a classic trap for cyclical semiconductor stocks. While Nvidia's 23.6x forward multiple looks attractive compared to the S&P 500, this assumes the current 97% revenue growth rate is a permanent state rather than a peak cycle. The market is pricing in a 'terminal' growth rate that ignores the inevitable deceleration as hyperscalers move toward internal silicon (ASICs) and the supply-demand imbalance normalizes. Investors are conflating 'cheap' with 'value' when the real risk is a contraction in gross margins as Blackwell production ramps up and competitive pressure from AMD and custom silicon intensifies. The stock isn't getting cheaper; it’s maturing.
If Nvidia continues to maintain its software moat via CUDA and cycles its product releases faster than competitors can iterate, the market may never force the multiple compression typical of standard hardware cycles.
"NVDA's forward valuation is only attractive if the company sustains 80%+ EPS growth through 2027—a bet the article makes implicitly but never stress-tests."
The article's core thesis—that forward estimate revisions post-earnings make NVDA cheaper on a valuation basis—is mechanically sound but obscures a critical risk: the article assumes the 97.2% revenue growth forecast is achievable and that AI capex cycles won't plateau. NVDA trades at 23.6x FY2027 EPS, which looks reasonable only if the company sustains mid-80s% growth through 2027. The SpaceX/Rubin tailwind is real but already priced into consensus. The bigger question: what if Wednesday's guidance suggests deceleration in H2 2025, or if data center customers optimize utilization rather than expand capacity? The article conflates 'cheaper valuation' with 'better investment'—they're not synonymous if growth disappoints.
If NVDA guides to sub-80% growth in FY2026 or signals customer inventory normalization, consensus EPS estimates will compress faster than they've expanded, and the 23.6x multiple becomes a value trap, not a bargain.
"Persistent AI capex and rising forward estimates could drive a re-rate for Nvidia, but the outcome hinges on ongoing demand strength and favorable macro conditions."
NVDA's earnings bounce is likely to lift sentiment, but the article leans on forward estimates as a perpetual fuel for a cheaper multiple. The strongest counterpoint is that forward revisions have already accelerated, meaning much of the growth is priced in and any near-term AI capex pause could trigger a multiple re-rating risk just when the headline numbers beat. The piece glosses over potential accelerant risks such as data-center demand normalization, margins peaking, and regulatory/export controls that could blunt growth. Also, the Rubin/SpaceX chatter, if speculative, could fade and leave investors relying on a fragile forecast. Still, if AI spending persists, the stock could re-rate on higher confidence in earnings power.
The strongest counterpoint is that Nvidia's growth could decelerate if AI demand cools or if policy/regulatory hurdles bite; a miss or weaker-than-expected guidance could trigger rapid multiple compression despite any near-term beat.
"Estimate revisions alone are unlikely to prevent multiple compression if AI capex growth decelerates in 2026."
The article correctly notes that NVDA forward EPS estimates have risen after each of the last four reports, leaving the stock at 23.6x fiscal-2027 EPS versus 97% expected Q2 revenue growth. Yet it underplays two structural risks: export curbs on China and the shift by Microsoft, Google, and Amazon toward custom ASICs that could cap Nvidia's data-center share after 2026. If Rubin ramps slower than hoped or gross margins slip below 75%, the re-rating the piece assumes may not materialize even if estimates tick up modestly.
A clean beat plus raised Rubin guidance could still drive another 15-20% EPS lift, pushing the multiple back toward 30x and rewarding holders regardless of near-term share-price reaction.
"Software-defined margin resilience will offset hardware commoditization as hyperscalers move toward internal ASICs."
Claude and Grok are fixated on the 'capex plateau,' but both ignore the software-defined margin expansion. Nvidia isn't just selling silicon; they are locking in recurring revenue via NVIDIA AI Enterprise. Even if hardware growth decelerates to 40% as ASICs proliferate, the software attach rate could sustain gross margins above 75%. The real risk isn't just 'capex fatigue'—it's the potential for a regulatory 'black swan' in the sovereign AI space that complicates global deployment.
"Software attach can't mathematically offset hardware share loss if hyperscalers internalize 40%+ of AI inference silicon by 2027."
Gemini's software-margin thesis is underexplored but needs stress-testing. NVIDIA AI Enterprise attach rates are real, but enterprise software margins (typically 70-80%) don't offset hardware margin compression if data-center ASICs capture 30%+ share by 2027. The regulatory black swan is valid; export controls already cost ~$5B annually. Neither risk is priced into consensus if 97% growth assumes unfettered China access and zero ASIC cannibalization.
"Software margins won’t fully offset hardware-margin compression; margin trajectory depends on a slower software attach, higher costs, and regulatory tail risks."
Gemini's software-margin thesis is intriguing but under-specified. Even with NVIDIA AI Enterprise, incremental software gross margins may not fully offset hardware-margin compression if data-center ASIC competition grows to 30%+ share by 2027, and enterprise maintenance plus sales costs erode the uplift. The risk isn’t just capex fatigue but a slower software attach ramp, higher opex, and potential customer concentration. Regulatory constraints add a low-probability, high-impact tail risk that the market currently underprices.
"Export controls will hasten ASIC adoption by giving hyperscalers extra reason to avoid Nvidia dependency."
Claude notes the $5B export-control cost but misses how those curbs directly accelerate ASIC development: hyperscalers already facing China restrictions have stronger incentive to build internal silicon to guarantee supply and bypass future curbs. If sovereign-AI contracts draw similar scrutiny, the 30% ASIC share threshold could hit by late 2026 rather than 2027, outpacing any NVIDIA AI Enterprise margin offset.
The panel consensus leans bearish, with concerns about NVDA's growth sustainability, potential margin compression due to ASIC competition, and regulatory risks outweighing the bullish case for software-defined margin expansion and AI enterprise growth.
Software-defined margin expansion via NVIDIA AI Enterprise
Margin compression due to ASIC competition and potential regulatory black swans in the sovereign AI space.